The Unit on Neuroimaging has continued to focus on studies of a simple form of learning and memory, called eyeblink conditioning. In our original study of this paradigm, using positron emission tomography (PET), we noted that many of the changes in regional cerebral blood flow seen were lateralized to one side of the brain. We hypothesized that this lateralization was a function of the side of the body to which the unconditioned stimulus was administered. To test this, we administered the unconditioned stimulus to the side opposite that from the original study. We found that for the most part, areas in the brain involved in learning also switched sides. This indicated that the areas involved in learning, such as the cerebellum and the primary auditory cortex, were not unique to the right or left cerebral hemisphere, as would be for example, language functions. In another arm of the project, we evaluated the effect of drugs known to affect learning and memory on eyeblink classical conditioning. The anti- cholinergic drug scopolamine has been used extensively in other paradigms to model aspects of the memory impairment that occurs in Alzheimer's disease and in normal aging. Lorazepam, a commonly prescribed benzodiazepine, also affects other memory tasks. Both drugs impaired learning of the conditioned response, as compared with an active control drug. An interesting differentiating point between the two drugs was that subjects on scopolamine failed to extinguish or "unlearn" the response under the appropriate conditions. We hypothesized that this was due to a disruption of hippocampal function, leading to failure to inhibit inappropriate responses, as can happen in patients with dementia. The learned association after lorazepam administration appeared to be more fragile, as demonstrated by quicker extinction of the learned response. This is consistent with other data on impairment of memory consolidation with benzodiazepines.